Printer Friendly

An integrated system of vapor-compression chiller and absorption heat pump for efficiency improvement: system modeling and performance analysis.


Buildings account for 41% and 35% of the primary energy use and greenhouse gases emissions, respectively. Among the 41% of the primary energy, 40% are used for building heating, cooling, and domestic hot water (DOE, 2011) in United States. Vapor-compression chillers (VC) have been widely used for space cooling. As one of approaches to improve the efficiency of the VCs (Minh et al., 2006), the dedicated mechanical subcooling system combines two VCs: a main chiller and another chiller with a smaller capacity through a heat exchanger, called a subcooler. The subcooler is actually the evaporator of the smaller chiller and it is located between the condenser and the expansion valve of the main chiller. The refrigerant of the small chiller is used to subcool the refrigerant of the main chiller in the subcooler. The refrigerant of the main chiller after the subcooler is at much lower temperature and lower quality before entering the evaporator. Thus more heat in the evaporator can be absorbed and a higher efficiency of the overall system can be achieved. There exist some investigations on the dedicated mechanical subcooling system. Couvillion et al. (1988) developed models for the dedicated mechanical subcooling system and they concluded that the improvements of COP and the system capacity were 80% and 170% respectively. Thornton et al. (1994) explored the optimal evaporation temperature of the smaller chiller to maximize the COP of the overall system. However, since both chillers utilize electricity as driven power in the two studies, the COP of the dedicated mechanical subcooling system is always between the COP of the main chiller and the smaller chiller, which is typical higher than the main chiller. Therefore the upper bound of the overall COP of the dedicated mechanical subcooling system based on two electrical VCs is the COP of the smaller VC.

Two studies investigated the performance improvement by using non-electrical driven chillers to replace the smaller vapor-compression chiller in the dedicated mechanical subcooling system. Hwang (2004) explored a system which used an absorption chiller driven by the exhaust gas from a microturbine to replace the smaller VC. The chilled water from the absorption chiller was used to subcool the refrigerant coming out from the condenser in the main chiller. Their results showed the COP was improved from 1.43 -- 3.70 to 1.78 -- 3.84 and the annual energy consumption was reduced by 12%. The simple payback period for the system was 7 years. Another system also used absorption chiller to replace the smaller chiller. The system was installed in San Antonio, Texas in 2005. According to the final report to Oak Ridge National Laboratory (2005), the system primarily consisted by an 18 tons absorption chiller driven by the exhaust gas from a 60 kW microturbine and a refrigeration system with the peak load of 130 tons. Although the overall complicated system was not cost effective because of high parasitic electric power consumption, and relatively high natural gas price, the operation data indicated that the subcooling technology was only one, which can achieve better performance and cost effectiveness in the couple of technologies tested in the project. Kung et al. (2013) proposed an integrated system in which an absorption heat pump is used to replace the smaller chiller to subcool the vapor-compression chiller. The results of their model show that the integrated system can have improve the COP of the vapor-compression chiller by 66%. Additionally, the integrated system can provide heating simultaneously.

A test bed now is being constructed according to the system proposed by Kung et al. (2013). This paper focuses on the thermodynamic and heat transfer model developed for the test bed and performance analysis. The models are able to predict the performance of the test bed at different working conditions and to compare the performance improvement with the VC alone.


The integrated system in the effort is named the Vapor-compression chiller and Absorption heat pump Integrated System (VAIS). The VAIS has been designed for typical building heating, cooling, and domestic hot water system. The test bed of the VAIS consists of a water-cooled VC, a natural gas driven absorption heat pump, and a heat exchanger. The test bed is now being constructed in the Herrick Lab at Purdue University, West Lafayette, IN. The VC in the test bed is a used Dunham-Bush water-cooled chiller driven by electricity, and the absorption heat pump (AHP) is a single-stage ammonia/water AHP from Robur driven by thermal energy. The VC and the ABS are integrated by a heat exchanger called the subcooler. The VC in the integrated system is designed as a chiller to provide cooling while the AHP is designed as a heat pump to provide domestic hot water or heating. The chilled water from the evaporator of the AHP is used to provide subcooling for the VC. The subcooler is installed between the condenser and the expansion valve in the VC. It uses the chilled water generated by the evaporator of the AHP to subcool the refrigerant exiting the condenser of the VC. This allows the refrigerant entering the evaporator with a lower quality and absorbing more heat. Therefore, comparing to the VC alone, the integrated system can use a smaller VC which requires less electricity to achieve the same cooling capacity.

Figure 1 is the schematic diagram of the VAIS test bed. The VC has four major components in series: a compressor (COM), a condenser (CON), an evaporator (EVA), and an expansion valve (EVA). Work is required to drive the compressor which circulates the refrigeration in the VC. The chilled water cooled by the evaporator is used for building cooling. In the AHP, the weak ammonia-water solution absorbs the mixture vapor at lower temperature and pressure in the absorber and become strong solution. The strong ammonia-water solution is heated at high temperature and pressure in the desorber by the external heat added. Ammonia with a little water vapor is boiled off from the strong ammonia-water solution in the desorber and rectifier (REC) and goes into the condenser. This AHP recovers internal heat intensively. For example, the working fluid exchanges heat at the precooler (PRE). Also, the strong ammonia-water solution from the absorber first ejects heat in a heat exchanger then absorbs heat back in the rectifier and the absorber. This design can increase the efficiency and reduce the size of the machine because the rectifier and the absorber do not need external cooling water connections. However, this operation cycle increases the complexity of modeling. The hot water return is heated by both the condenser and another heat exchanger, which gains the heat from the rectifier and the absorber in the AHP, as shown in Figure 1.



Thermodynamic and heat transfer models of the VAIS have been developed in Engineering Equation Solver (EES), developed by University of Wisconsin. EES can provide the numerical solution of a set of algebraic equations. EES' large built-in data bank of thermodynamic and transport properties are helpful in solving problems in thermodynamics, fluid mechanics, and heat transfer and it is particularly useful for design problems. The assumptions of the VAIS models developed include:

1. Pressure drops along the pipes are neglected.

2. Heat losses of the components and along the pipes are neglected.

3. Working fluids exiting the condensers of both the VC and the ABS are saturated liquids.

4. Weak solution exiting the desorber is saturated liquid.

5. Working fluids exiting the desorber and rectifier of the AHP are saturated vapors.

6. Mass fraction of the working fluid exiting the rectifier is assumed as 99%.

In the real machine of the AHP, three streams exchange heat in a heat exchanger called condenser/absorber. These three streams are the refrigerant vapor, the strong solution, and the hot water return. In the developed model, the condenser/absorber was separated into two independent heat exchangers and linked by an additional exchanger marked HX as shown in Figure 1. The refrigerant vapor and the strong solution both exchange heat with the hot water in parallel and mix together afterwards. In this configuration, it is assumed the condenser and the heat exchanger, HX at same temperatures but different flow rates. The two separated hot water streams will combined together again when they exit the two heat exchangers. This change basically refers to other similar studies (Lazzarin et al., 1996; Horuz et al, 2003; Darwish et al, 2008; Rossa and Bazzo, 2009). It could simplify the models and is relatively close to the real situation.

Heat Exchangers

The most components in the VAIS are heat exchangers. The NTU-[epsilon] method was used for analysis of all the heat exchangers in the study. The NTU-[epsilon] method uses the effectiveness, e, of the heat exchanger and the inlet states of the hot and cold fluids to calculate the heat transfer rate. The effectiveness of a heat exchanger is the ratio of the actual heat transfer rate, [[??].sub.act], and the maximum possible heat transfer rate, [[??].sub.max], as expressed in Equation 1.

[epsilon] = [[??].sub.act]/[[??].sub.max] (1)

The maximum possible or the theoretical heat transfer rate of a heat exchanger is the product of the temperatures difference between the inlets of the hot and cold fluids, and the smaller value between the products of the mass flow rate and the specific heat of the hot and cold fluids, respectively. The related equations are expressed in Equations 2-5.

[[??].sub.max] = [C.sub.min] x ([T.sub.h,i] - [T.sub.c,i]) (2)

[C.sub.min] = min([C.sub.h], [C.sub.c]) (3)

[C.sub.h] = [[??].sub.h] x [C.sub.p,h] (4)

[C.sub.c] = [[??].sub.c] x [C.sub.p,c] (5)

[[??].sub.act] = [C.sub.h] x ([T.sub.h,i] - [T.sub.h,o]) (6)

[[??].sub.act] = [C.sub.h] ([T.sub.h,i] - [T.sub.h,o]) (7)

The actual heat transfer rate of a heat exchanger is calculated by the products of mass flow rate, specific heat, and the temperature difference of the hot and cold fluids as shown in Equation 6 and 7. Once the effectiveness of the heat exchanger, the inlet temperatures and flow rates of the hot and cold fluids are given, the theoretical and actual heat transfer rate of a heat exchanger can be calculated. The outlet temperatures of the hot and cold fluids can be calculated.


Isentropic efficiency is used for the compressor modeling in the VC. The isentropic efficiency is defined as the ratio of the enthalpy difference of isentropic process and the enthalpy difference of real process across the compressor. As shown in Equation 8. A more comprehensive compressor model based on physical and working principle will be added and replace the current model given the more detailed information of the VC is available.

[[eta].sub.isen (8)] = [h.sub.o(so=si)] - [h.sub.i]/[h.sub.o] - [h.sub.i] (8)

Expansion Valves and Pumps

All the expansion valves work at isenthalpic condition in the study. The enthalpies of the working fluids entering and exiting the expansion valves are the same.

The power consumption of the solution pump in the AHP can be calculated by the products of the mass flow rate, the inlet specific volume, and the pressure difference of the passing working fluid. However, the power consumption of the solution pumps in the AHP is negligible because the influence of them is very little.

Other inputs for the models

In the VAIS models, some parameters were set as input according to given information from manufactory. In the VC, the refrigerant exiting the evaporator of the VC is set at 4.44[degrees]C superheated and the flow rate of the cooling water provided to the condenser of the VC is 3gpm/ton at 85[degrees]F (29.44[degrees]C). Based on rated condition, the chilled water provided from the VC and AHP was set at 7[degrees]C and the hot water return temperature was set at 40[degrees]C.


Rated Condition

The VAIS models developed are able to calculate the working condition and system performance. The follow section uses the calculated model results of a VAIS with the capacity of 68.75 kW to address the system performance predicted. In order to easily read data, the working conditions of the VAIS are summarized in Table 1 based on the calculation results of the VAIS model at rated condition. The working conditions include pressure, P; temperature, T; enthalpy, h; mass fraction, x; and mass flow rate, m of the working fluids in the VAIS. The state points in Table 1 are indicated in Figure 1.


The performance of the VAIS was compared with the VC with the same capacity to address the efficiency improvement achieved by the VAIS, as shown in Table 1. The VAIS could reduce the power consumption from 14.59 kW to 12.23 kW to provide the same cooling capacity at 68.75 kW. Thus, the COP for cooling of the VAIS is increased from 4.71 to 5.62, which indicates a 19.32% improvement. Additionally, the VAIS can produce 39.15 kW of hot water at 51.04[degrees]C. The hot water can be directly used as the domestic hot water or spacing heating with addition heat inputs. Based on the modeling results, the VAIS can reduce the electricity consumption by 15-20% and increase the efficiency by approximant 20%.

Except overall system performance, the system sensitivity has been studied. First of all, if the cooling water inlet temperature, T[30], changes, the performances of the VC and the VAIS changes accordingly. As shown in Figure 2(a), the COPs decreases as the cooling water temperature inlet increases in both the VC and VAIS. This is because higher the condenser temperature, less condensation is produced. However, COP improvement, which the difference between the COPs of the VC and the VAIS, increases as the cooling water temperature increases as shown in Figure 2(b). This means the VAIS can more effectively improve the efficiency when the VC performs poorly. This makes agreement with the previous study (Couvillion et al., 1988).

Additionally, Figure 3 shows how the heat transfer rates of the subcooler and the condenser of the VC change when the cooling water inlet temperature changes. Since the cooling capacity is fixed, these two heat transfer rates should act in the opposite ways as the figure shows. When the inlet cooling water temperature is low, it is possible to remove more heat at the condenser and on the other hand, the heat duty of the subcooler decreases.

The capacity of the subcooler is equal to the cooling capacity of the AHP. Once the heat transfer rate of the subcooler decreases, the heat transfer rate of the evaporator of the AHP also decreases. This results in the decrease of the heating capacity of the AHP in the VAIS. Figure 4(a) shows how the heating capacity decreases as the cooling water inlet temperature decreases. The hot water supply temperature basically follows the trend of the heating capacity as Figure 4(b) shows. However, the hot water temperature is still maintained around 50C .


Compared to the VC alone, the VAIS is able to utilize less power for the VC to achieve the same cooling capacity. If the AHP in the VAIS utilizes free thermal energy such as waste heat or hot fluid from solar energy, the COP of the VAIS can be significantly increased. According to the model calculation results, the VAIS is possible to improve the COP from 13% to 28% depends on the operation condition when the AHP in the VAIS is driven by free thermal energy. The VAIS can have higher improvement if the VC works when the cooling water inlet temperature is higher. The cooling and heating performances of the AHP are also affected by cooling water condition. If more heat can be removed from the condenser of the VC, the cooling heat duty of the AHP decreases and so does the heating production. The hot water supply temperature is also affected although it approximately maintain around 50[degrees]C. The calculation results from the developed models will be used to compare with the experiment data collected soon and to conduct system sensitivity analysis and optimization. In summary, according to our calculations, the VAIS studied has a great potential to reduce energy consumption and greenhouse gases emissions by at least 15%. The VAIS also provides more opportunities to integrate the waste heat and renewable energy systems with the conventional cooling system for higher efficiencies.





C = product of mass flow rate and specific heat, kj/s[degrees]C act = actual

[C.sub.p] = specific heat, kJ/kg[degrees]C c = cold fluid

h = enthalpy, kJ/kg h = hot fluid

[??] = mass flow rate, kg /s [ = inlet

[??] = heat transfer rate, kW isen = isentropic

T = temperature, [degrees]C

Greek symbols

[epsilon] = effectiveness of heat exchanger

[eta] = efficiency


act = actual

c = cold fluid

h = hot fluid

i = inlet

isen = isentropic

max = maximum

min = minimum

o = outlet

so = entropy of outlet

si = enrtopy of inlet

max = maximum

min = minimum

o = outlet

so = entropy of outlet

si = entropy of inlet


Couvillion, R.J., Larson, M.W., and M.H. Somerville. 1988. Analysis of a vapor-compression refrigeration system with mechanical subcooling. ASHRAE transactions, 94, 641-660.

Darwish, N. A., Al-Hashimi, S. H., and A.S. Al-Mansoori. 2008. Performance analysis and evaluation of a commercial absorption--refrigeration water--ammonia (ARWA) system. International Journal of Refrigeration, 31(7): 1214-1223.

DOE. 2011 Buildings Energy Data Book. Available June 14, 2013, from

Horuz, I., and T.M.S. Callander. 2004. Experimental investigation of a vapor absorption refrigeration system. International journal of refrigeration, 27(1): 10-16.

Kung, Y.S., Qu, M., and S. Peng. 2013. Model based analysis of an integrated system of vapor-compression chiller and absorption heat pump. International Conference on Energy Sustainability.

Lazzarin, R. M., Gasparella, A., and G.A. Longo. 1996. Ammonia-water absorption machines for refrigeration: theoretical and real performances. International journal of refrigeration, 19(4): 239-246.

Minh, N.Q., Hewitt, N.J., and P. C. Eames. 2006. Improved Vapour Compression Refrigeration Cycles: Literature Review and Their Application to Heat Pumps. International Refrigeration and Air Conditioning Conference, Paper 795.

NAEA. 2005. Final Report to Oak Ridge National Laboratory Under Subcontract Number 4000016141

Rossa, J. A., and E. Bazzo. 2010. Thermodynamic modeling of an ammonia-water absorption system associated with a microturbine. International journal of thermodynamics, 12(1): 38-43.

Thornton, J.W., Klein, S.A., and J.W. Mitchell. 1994. Dedicated mechanical-subcooling design strategies for supermarket applications. International Journal of Refrigeration, 17 (8): 508-515.

Y. Hwang. 2004. Potential energy benefits of integrated refrigeration system with microturbine and absorption Chiller. International Journal of Refrigeration, 27: 816-829.

Yi-Shu Kung

Student Member ASHRAE

Ming Qu, PhD

Associate Member ASHRAE

Steve Peng, PhD

Yi-Shu Kung is a Ph.D. student at Purdue University, West Lafayette, IN. Ming Qu is an Assistant professor in School of Civil Engineering, Purdue University, West Lafayette, IN. Steve Peng is an Associate professor in California State University, East Bay, Hayward, CA.
Table 1. Properties of the working fluids in the VAIS

 P (kPa) T ([degrees]C) h (kJ/kg) x m (kg/s)

1 1847 102.4 304.1 0.4644 0.0440
2 1847 159.9 570.2 0.1873 0.0288
3 508.2 117.2 570.2 0.1873 0.0288
4 1847 96.24 1472 0.9758 0.0156
5 1847 96.24 198.8 0.4644 0.000422
6 1847 81.41 1410 0.99 0.0152
7 1847 46.71 217 0.99 0.0152
8 1178 30.67 217 0.99 0.0152
9 1178 16.04 67.79 0.99 0.0147
10 508.2 4.871 67.79 0.99 0.0147
11 508.2 5.658 965.7 0.99 0.0147
12 508.2 6.687 1120 0.99 0.0147
13 508.2 75.64 454.3 0.458 0.0435
14 508.2 46.71 -29.34 0.458 0.0435
15 1847 46.84 -27.71 0.458 0.0435
16 1847 54.52 6.281 0.4644 0.0440
17 101.3 40 167.5 -- 0.9240
18 101.3 40 167.5 -- 0.3925
19 101.3 51.04 213.7 -- 0.3925
20 101.3 40 167.5 -- 0.5315
21 101.3 49.46 207.1 -- 0.5315
22 101.3 50.13 209.9 -- 0.9240
23 101.3 11.12 46.7 -- 0.7620
24 101.3 7 29.42 -- 0.7620
25 591.5 10.39 410.8 -- 0.3510
26 1576 73.03 445.6 -- 0.3510
27 1576 41.11 252.4 -- 0.3510
28 1576 12.36 214.9 -- 0.3510
29 591.5 5.393 214.9 -- 0.3510
30 101.3 29.44 123.3 -- 2.85
31 101.3 35.13 147.1 -- 2.85
32 101.3 13.52 56.72 -- 2.518
33 101.3 7 29.42 -- 2.518

Table 2. Comparison of the system performances
between the VAIS and the VC

Performance VC VAIS

Cooling capacity (kW) 68.75 68.75
Power consumption (kW) 14.59 12.23
Cooling COP 4.71 5.62
COP improvement (%) -- 19.32
Heating capacity (kW) -- 39.15
Hot water supply -- 51.04
COPYRIGHT 2014 American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kung, Yi-Shu; Qu, Ming; Peng, Steve
Publication:ASHRAE Transactions
Article Type:Report
Geographic Code:1USA
Date:Jan 1, 2014
Previous Article:Aerogel-coated metal foams for dehumidification applications.
Next Article:Field demonstration of a cold climate heat pump.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters